28105 A. Moosavian et al./ Elixir Control Engg.

76 (2014) 28105-28108

Available online at www.elixirpublishers.com (Elixir International Journal)

Control Engineering
Elixir Control Engg. 76 (2014) 28105-28108

Journal-bearing fault detection based on vibration analysis using feature

selection and classification techniques
A. Moosavian, H. Ahmadi and A.Tabatabaeefar
Department of Mechanical Engineering of Agricultural Machinery, University of Tehran.
A R TI C L E I N F O A B ST R A C T
Art i c l e h i st ory : Vibration technique in a machine condition monitoring provides useful reliable information,
Received: 28 May 2012; bringing significant cost benefits to industry. By comparing the signals of a machine running
Received in revised form: in normal and faulty conditions, detection of faults like journal-bearing defects is possible.
25 October 2014; This paper presents an appropriate procedure for the fault detection of main engine journal-
Accepted: 31 October 2014; bearing based on vibration analysis. The frequency-domain vibration signals of an internal
combustion engine (IC engine) with normal and defective main journal-bearings were
K ey w or d s obtained. The signal processing technique plays one of the important roles for recognizing
Fault detection, the journal-bearing fault in the proposed system. In the present research, the data mining
Vibration analysis, method based on feature extraction and selection is proposed. The database is established by
Journal-bearing of internal combustion the feature vectors of frequency domain signals which are used as input pattern in the
engine, training and identification process. The SVM and KNN is proposed to identify and classify
Signal processing, the journal-bearing fault conditions in the condition monitoring system. The experimental
Data mining, results verified that the proposed diagnostic procedure has more possibilities and abilities in
Support vector machine, the fault diagnosis of the main journal-bearing of IC engine.
K-nearest neighbor. © 2014 Elixir All rights reserved.

Introduction Vibration analysis in particular has been applied for some

Since the condition monitoring has significant impacts in
industry, it has received an enormous attention from the expert
and practical maintenance. According to study, maintenance
costs are a major part of the total operating costs of all
manufacturing, which can make or break a business [1].
Reliability has always been an important aspect in the
assessment of industrial products [16]. With the increase in
production capabilities of modern manufacturing systems, plants
are expected to run continuously for extended hours. Therefore,
the condition monitoring of machines, especially early fault
diagnosis, is proved to be necessary and has been received wide
attentions in this decade [17]. By development of technology,
cost of time-based preventive maintenance increased thus, new
approaches in maintenance such as condition-based maintenance
(CBM) developed [18].
Most of machinery used in the modern world operates by
means of rotary components which can develop failures. The
monitoring of the operative conditions of rotating machinery
provides a great economic improvement by decreasing
maintenance costs, as well as improving the safety level. So, it is
essential to analyze the external information so as to evaluate the
internal components state which, generally, are inaccessible
without disassemble the machine [19].
Fault diagnosis is a wide and active area of research. There
are a large volume of articles that deal with this subject [2]. In
many applications the problem of fault diagnosis is an important
issue that has been theoretically and experimentally investigated
with different types of methods. Fault diagnosis improves the
reliability and availability of an existing system. Since various
failures degrade relatively slowly, there is potential for fault
diagnosis at an early step [3].
time as a predictive maintenance method and as a support for
machinery maintenance decisions [20]. Vibration analysis
provides the most information from the data acquired [5].
Vibration monitoring of rotary machines has become an
appealing field for many researchers and has also achieved
industrial acceptance [22].
Journal-Bearings are multifunctional devices. In order to
operate efficiently and provide long service life, journal-
bearings often have to satisfy
several requirements simultaneously. These include:
- Position and support a crankshaft or journal and permit motion
with minimum energy consumption;
- Support a fixed load and be able to withstand occasional shock
loads;
- Run quietly and suppress externally generated vibrations;
- Resist corrosion under normal service conditions as well as
during storage or extended down-time [6].
A crankshaft spinning within a journal bearing is actually
separated from the journal bearing’s metal facing by an
extremely thin film of continuously supplied engine oil that
prohibits metal to metal contact (Figure 1).
Figure 1. The mechanism of crankshaft motion in
journal-bearing

Tele:
E-mail addresses: a.moosavian@ut.ac.ir
© 2014 Elixir All rights reserved
28106 A. Moosavian et al./ Elixir Control Engg. 76 (2014) 28105-28108

Journal bearings are applied to carry radial loads, for The sequence of N complex numbers x0,...,xN−1 is
example, to support a crankshaft of engine. Journal bearings are transformed into the sequence of N complex numbers X0,...,XN−1
considered to be sliding bearings as opposed to rolling bearings by the DFT according to the formula:
such as ball bearings. Despite this categorization, a crankshaft N −1 −
2π i
kn , k = 0,..., N − 1 (1)
spinning within a journal bearing is actually separated from the X k = ∑ xn e N

journal bearing’s metal facing by an extremely thin film of n =0

continuously supplied engine oil that prohibits metal to metal
contact. As such, the journal bearing allows the crankshaft to
normally be contacted only by oil, which explains the long life
of engines that get regular oil changes. So, the detection and
understanding of condition degradation in the journal-bearing is
important for engine performance [21].
In this study, a procedure was performed to the detection of
journal-bearing fault using vibration analysis. Also, the
performance comparison of SVM and KNN was done. Two
conditions of the journal-bearing are studied, namely, normal
and faulty. The effect of FFT, feature extraction and selection
techniques, SVM and KNN classifiers was surveyed. Figure 2
shows flow diagram of the proposed procedure.
Experimental Engine
The experimental system for this work is an internal
combustion engine. The working speed of its crankshaft is set at
1500 rpm. Vibration signals were measured for normal and fault
conditions by an accelerometer sensor (VMI-102 model) which
was mounted horizontally on the crankcase of engine. The
frequency-domain signals were gained by X-Viber data
acquisition with 8192 Hz sampling rate. The time was 4 second
for each sampling. The number of data in each sample was
12800. In this paper, we studied the vibration behavior of the IC
engine for normal and defective journal-bearings. Then we
estimate the performance of data mining and SVM on
classifying the train and test dataset. MATLAB software was
used for this work.
Data Mining
Feature Extraction
30 features are extracted from time and frequency-domain
signals included Maximum, Minimum, Average, Root Mean
Square (RMS), Standard Deviation (Stdv), Variance (Var), 4th
Momentum, 5th momentum, Crest Factor, Skewness, Kurtosis,
etc.
Feature Selection
Stepwise selection is a method that allows moves in
direction, dropping or adding variables at the various steps.
Forward stepwise selection involves starting off in a forward
approach and then potentially dropping variables if they later
appear to be not significant. This means that starting with empty
the variable set, stepwise forward selection add the variable
which leads to the smallest decrease in prediction error.
Significance is measured by a partial F-test. In this work the
value of p (significance level) is 0.2. In this step 2 features are
selected [24].
Support vector machine
The Support Vector Machine (SVM) has been developed by
Vapnik and is gaining popularity due to many appealing
features, and promising empirical performance.SVM is a
popular technique for machine learning activities including
classification and regression. The power of SVM in
classification leads to abundant application of this in fault
detection. SVMs are introduced on the foundation of statistical
learning theory [11-12-13]. Support vector machines (SVM)
were originally designed for binary (2-class) classification. The
SVM tries to place a linear hyperplane between the two different
classes [14-15].
The hyperplane can be expressed in following terms:
n , x = {1, 2,...n} , w ∈ R N ,
wT .x + b = ∑ w j x j + b = 0
j =1

b∈R (2)
where the vector w defines the hyperplane, b is a scalar
threshold. The equations for class 1 and -1, respectively, are
( w.x ) + b = 1 (3)
and
( w.x) + b = −1 (4)
The following decision function holds good for all data points
Figure 2. Diagram of diagnostic procedure belonging to either 1 or -1:
Signal Processing f ( x ) = sign (( w.x) + b) (5)
The great advances in vibration analysis in recent years are
the extension in signal processing techniques, for vibration The optimal hyperplane can be obtained by solving the
diagnostics of journal-bearing defects [6-7-8-9]. The frequency optimization problem:
domain spectrum is more useful in identifying the exact nature minimize: 1 2 (6)
τ ( w) = w
of fault in the journal-bearings. Although there are various 2
techniques, the analysis of vibration signals is often based on the subject to: yi (( w.xi ) + b) ≥ 1 , i = 1,..., l (7)
Fast Fourier Transform (FFT) [10, 23]. where l is the number of training sets. The solution of the
The FFT is a faster version of the Discrete Fourier
Transform (DFT). The Fourier transform converts waveform constrained quadratic programming (QP) optimization problem
data in the time domain into the frequency domain. The Fourier can be obtained as
transform implements this by breaking down the original time- w = ∑ vi xi (8)
based waveform into a series of sinusoidal terms, each with a
unique magnitude, frequency, and phase.
28107 A. Moosavian et al./ Elixir Control Engg. 76 (2014) 28105-28108

where x are SVs obtained from training. Putting (8) in (5), the
i
decision function is obtained as follows:
l (9)
f ( x ) = sign(∑ vi ( x.xi ) + b)
i =1
In cases where the linear hyperplane in input spaces will not be
enough to separate two classes exactly, it is possible to construct
a hyperplane that allows linear separation in the higher Figure 3. Frequency spectrum of normal journal-
dimension (feature space). In SVMs, this is achieved through the bearing
use of a mapped function φ ( x ) that transforms the data from an
N -dimensional input space to Q -dimensional feature space.
Substituting the mapping in (9) gives
l (10)
f ( x) = sign(∑ vi (φ ( x ).φ ( xi )) + b)
i =1
The mapping into higher-dimensional feature space is relatively
severe computation. A kernel can be used to perform this Figure 4. Frequency spectrum of defective journal-bearing
mapping. It means that the transformation can be replaced by an The data mining process is significantly effective in
equivalent kernel function. This helps in reducing the classification results. 30 features were extracted from the
computational load. The kernel function K ( x, y ) is defined as vibration signals in frequency-domain. Then 2 superior features
K ( x, y ) = φ ( x ).φ ( y ) (11) were selected by the Stepwise Backward Selection technique
The decision function is accordingly modified as with p value of 0.05 and linear model. The dataset consisting of
l (12) 2 features and 40 samples were divided into two subsets,
f ( x) = sign(∑ vi K ( x, xi ) + b) training set and test set. The SVM classifier was trained by
i =1 training dataset, and then their performance was exactly
There are different kernel functions like polynomial, quadratic, estimated by the test dataset.
sigmoid, RBF, multilayer perceptron. For SVM, the target values were specified as 1 and -1,
In below, RBF kernel, given by Eq. (13) is described [25]. respectively, representing normal and defective journal-bearing.
− x− y
2 , 0 < σ < +∞ (13) The SVM with RBF kernel width of 1 (σ) and QP method were
K ( x, y ) = exp( 2
) applied for this work. Figure 5 shows the classification result of
2σ journal-bearing conditions. The train and test success of the
For SVM, QP method and the RBF kernel with σ=1 are used. SVM was 95% and 93%. So, it can be found that the
K-Nearest Neighbor performance of proposed procedure was perfect in the fault
Some training samples are used for train the KNN rule. K diagnosis of main engine journal-bearing.
nearest neighbour rule holds position of training samples and
their class. When decision about new incoming data is needed,
distance between query data and training samples is being
calculated. Based on the defined threshold for the rule (it is the
K number), K samples with least distances is selected and the
class with more samples inbound is the result. In the other word,
for example if there is 2 or 3 features for a classification
situation, position of training samples and input sample can be
visualized on 2D and 3D
Cartesian coordinates. Process to find result is like to draw a
circle (Sphere) centred on input location and increase radius
until k samples are embed inside the circle (sphere) and then a
class with more samples inbound is the result. Figure 5. The classification of normal and defective journal-
Without prior knowledge, the KNN classifier usually bearing by using SVM
applies Euclidean distances as the distance metric. However, this KNN is a classifier that its accuracy is always 100% on
simple and easy-to-implement method can still yield competitive training dataset, because the KNN holds the position of training
results even compared to the most sophisticated machine dataset and their class during the classification process. The test
learning methods [26]. The Euclidean distance between point p success of the KNN just depends on the K value. The K value of
and q is the length of the line between them. 1 was used for the KNN. The performance of KNN was 89.84%.
Results and Discussion By attention to the results, it can be found that the
The aim of this paper is the journal-bearing fault diagnosis performance of SVM was significantly better than KNN in the
of engine. To fulfill this purpose, FFT signal processing, data fault detection of main engine journal-bearing.
mining technique, SVM and KNN classifiers were used. Conclusion
Figure 3 and 4 shows the spectrum analysis of the normal A procedure was presented for the fault diagnosis of main
and faulty journal-bearing. It is obvious that the overall vibration journal-bearing in IC engine by using the data mining technique
amplitude of journal-bearing for faulty condition is more. and Support vector Machine classifier. The Fast Fourier
Transform technique is applied for processing signals. The
features of the generator output signal at different journal-
28108 A. Moosavian et al./ Elixir Control Engg. 76 (2014) 28105-28108
bearing conditions are extracted by the statistical and vibration
parameters. To further diminish the dimensionality of the
extracted feature vectors and enhance the model performance,
stepwise backward selection was used. Finally, the feature
vectors are fed into the SVM and KNN. The performance of
SVM is calculated by applying RBF kernel and Quadratic
Programming method. KNN with K=1 and Euclidean distance
was used. The results of recognition rates are used to estimate
the work efficiency in the proposed system. The FFT and SVM
techniques can provide good diagnosis efficiency for fault
journal-bearing positions identification and journal-bearing fault
conditions classification, respectively. The total recognition
rates are over 93%. Therefore, the results indicated that the
proposed system can provide an accurate and automatic
diagnosis technique for the journal-bearing of IC engines.
Acknowledgement
The authors would like to thank Tehran University for its
support to this research.
References
[1] Widodo A, Yang BS. Support vector machine in machine
condition monitoring and fault diagnosis. Mechanical Systems
and Signal Processing 2007:21: pp 2560-2574.
[2] Huang B. Detection of abrupt changes of total lest square
models and application in fault detection. IEEE Transactions on
Control Systems Technology 2001;9(2): pp 357-367.
[3] Mollazade K, Ahmadi H, Omid M, Alimardani R. Vibration-
Based Fault Diagnosis of Hydraulic Pump of Tractor Steering
System by Using Energy Technique. Modern Applied Science
2009;3(6): pp 59-66.
[4] Selection Guide with Bound and Hydro Computer-Assisted
Sleeve Bearing Design. Cast Copper Alloy Sleeve Bearings.
Copper Development Association.
[5] Ebrahimi E. Vibration analysis for fault diagnosis of rolling
element bearings. Journal of American Science 2012;8(2): pp
331-336.
[6] Baydar N, Ball A. A Comparative study of acoustic and
vibration signals in detection of gear failures using Wigner-Ville
distribution. Mechanical Systems and Signal Processing
2001;15: pp 1091-1107.
[7] Rao MA, Srinivas J, Rama Raju VBV, Kumar KVS.
Coupled torsional-lateral vibration analysis of geared shaft
systems using mode analysis. Journal of Sound and Vibration
2003:261: pp 359-364.
[8] Andrade FA, Esat I, Badi MNM. A new approach to time
domain vibration condition monitoring: gear tooth fatigue crack
detection and identification by the Kolmogorov-Smirnov test.
Journal of Sound and Vibration 2001;240: pp 909-919.
[9] Ho D, Randall RB. Optimization of bearing diagnostic
techniques using simulated and actual bearing fault signals.
Mechanical System Signal Process 2000;14: pp763-788.
[10] Hariharan V, Srinivasan PSS. New approach of
classification of rolling element bearing fault using artificial
neural network. Journal of Mechanical Engineering 2009;40(2):
pp119-130.
[11] Gunn SR, Brown M, Bossley KM. Network performance
assessment for neurofuzzy data modeling. Intelligent Data
Analysis 1997;1208 of Lecture Notes in Computer Science: pp
313-323.
[12] Shirvastava SK, Gharde SS. Support Vector Machine for
Handwritten Devanagari Numeral Recognition. International
Journal of Computer Applications 2010;7(11).
[13] Shigeo A. Support Vector Machine for Pattern
Classification. Springer. 2005: pp 237-245.
[14] Vapnik V. Statistical Learning Theory. Springer, NewYork.
1998.
[15] Cortes C, Vapnik V. Support-vector network. Machine
Learning 1995;20: pp 273-297.
[16] Bagheri B, Ahmadi H, Labbafi R. Implementing discrete
wavelet transform and artificial neural networks for acoustic
condition monitoring of gearbox. Elixir Mech. Engg 2011;35: pp
2909-2911.
[17] Niu G, Widodo A, Son JD, Yang BS. Decision-level fusion
based on wavelet decomposition for induction motor fault
diagnosis using transient current signal, Expert Systems with
Applications 2008:35: pp 918-928.
[18] Khazaee M, Ahmadi H, Omid M, Moosavian A. Fault
diagnosis and classification of planetary gearbox of MF285
tractor final drive using Fast Fourier Transform (FFT), Stepwise
Backward Selection and support vector machine (SVM)
classifier. Elixir Control Engg 2012;43: pp 6974-6977.
[19] Ahmadi H, Moosavian A. Fault Diagnosis of Journal-
Bearing of Generator Using Power Spectral Density and Fault
Probability Distribution Function. Innovative Computing
Technology, Communications in Computer and Information
Science 2011;241(1): pp 30-36.
[20] Wowk V. Machinery vibration: measurement and analysis.
McGraw-Hill Publishers. 1991: pp 64-69.
[21] Mollazade K, Ahmadi H, Omid M, Alimardani R. An
intelligent combined method based on power spectral density,
decision trees and fuzzy logic for hydraulic pumps fault
diagnosis. International Journal of Intelligent Systems and
Technologies 2008;3(4): pp 251-263.
[22] Moosavian A, Ahmadi H, Tabatabaeefar A. Condition
monitoring of engine journal-bearing using power spectral
density and support vector machine. Elixir Mech. Engg 2012:43:
pp 6631-6635
[23] Ebrahimi E. Fault diagnosis of Spur gear using vibration
analysis. Journal of American Science 2012;8(1): pp 133-138.
[24] The LOGISTIC Procedure, Chapter 39, Overview, SAS
OnlineDocä;Version 8: pp 1903-2040.
[25] Samanta B, Al-Balushi KR, Al-Araimi SA. Artificial neural
networks and support vector machines with genetic algorithm
for bearing fault detection. Engineering Applications of
Artificial Intelligence2003;16: pp 657-665.
[26] Y.Song, et.al, IKNN: Informative K-nearest neighbor
pattern classification. Proc. Springer-Verlag Berlin Heidelberg,
LNAI 4702, 2007: pp 248–264.